keras-team/keras · error · NotImplementedError

`filepaths` property method has not been implemented in {}.

Error message

`filepaths` property method has not been implemented in {}.

What it means

The base Iterator class declares filepaths as an abstract property; only DirectoryIterator (and iterators backed by files) implement it. Accessing it on an iterator that has no notion of files (e.g. NumpyArrayIterator) raises NotImplementedError by design.

Source

Thrown at keras/src/legacy/preprocessing/image.py:374

                (len(batch_x), len(self.class_indices)), dtype=self.dtype
            )
            for i, n_observation in enumerate(index_array):
                batch_y[i, self.classes[n_observation]] = 1.0
        elif self.class_mode == "multi_output":
            batch_y = [output[index_array] for output in self.labels]
        elif self.class_mode == "raw":
            batch_y = self.labels[index_array]
        else:
            return batch_x
        if self.sample_weight is None:
            return batch_x, batch_y
        else:
            return batch_x, batch_y, self.sample_weight[index_array]

    @property
    def filepaths(self):
        """List of absolute paths to image files."""
        raise NotImplementedError(
            "`filepaths` property method has not "
            "been implemented in {}.".format(type(self).__name__)
        )

    @property
    def labels(self):
        """Class labels of every observation."""
        raise NotImplementedError(
            "`labels` property method has not been implemented in {}.".format(
                type(self).__name__
            )
        )

    @property
    def sample_weight(self):
        raise NotImplementedError(
            "`sample_weight` property method has not "
            "been implemented in {}.".format(type(self).__name__)

View on GitHub (pinned to 7a34a03db6)

Solutions

Use flow_from_directory(...) instead of flow(...) when you need file paths
Guard access with try/except NotImplementedError before reading the property
In custom Iterator subclasses, override filepaths to return the source paths

Example fix

# before
paths = iterator.filepaths  # raises on NumpyArrayIterator

# after
try:
    paths = iterator.filepaths
except NotImplementedError:
    paths = None

Defensive patterns

Strategy: try-catch

Validate before calling

paths = iterator.filepaths if type(iterator).__name__ == 'DirectoryIterator' else None

Type guard

def has_filepaths(it):
    try:
        it.filepaths
        return True
    except NotImplementedError:
        return False

Try / catch

try:
    paths = iterator.filepaths
except NotImplementedError:
    paths = None  # in-memory data has no files

Prevention

Feature-detect before introspecting iterators
Prefer DirectoryIterator when tooling needs image paths

When it happens

Trigger: Calling iterator.filepaths on the result of flow(x, y) (NumpyArrayIterator), or on a custom Iterator subclass that did not override the property.

Common situations: Generic code that introspects iterators (Grad-CAM explainability tools, error-analysis scripts) run against in-memory numpy data instead of a directory dataset.

Related errors

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/5e5de345e0e57d00. Report an issue: GitHub.